Secrets Of Success: With-profits: is it time to surrender?

Regular readers will know that I sustain an interest in the performance of with-profits policies, although the subject is one of the dullest imaginable. Every year around this time, when the life companies start to announce their latest bonuses, I summon the energy to check the surrender values of my two surviving policies. Originally the purpose was to check on their progress: more recently it's been to work out whether or not it is worth surrendering them.

This exercise has become a chore. The critical issue is not so much what the prospects for the policies achieving their given target final sums are, but whether the opportunity cost of continuing to pay the premiums is worth it. There may be much better choices elsewhere, even allowing for the potential terminal value. (In retrospect, of course, there always are superior alternatives: the issue is whether you can identify them in advance).

Given my argument that many investors would in general do better to take greater control of their own investments (it is not as scary as it might seem, and saves you from paying unnecessary amounts to fund the lifestyles of intermediaries who may know less about investment than you do), you may wonder why I still hold any policies at all.

In my case, the reason has nothing to do with the investment case, but everything to do with the small print, which makes surrendering policies liable to tax if you surrender them within 10 years of a "chargeable event". A chargeable event can arise, for example, if you divorce and retain a policy that was formerly held in joint names. It means that you are liable to capital gains tax on the realised gains in the policy if you surrender before the 10 years are up.

In my case, the 10-year period expires this year, so I have been studying the latest with-profits survey by the trade magazine Money Management with more than usual interest. It is an invaluable source, still available in some public libraries(although the dense tables get no easier to read).

A further reason for checking on the data is that, several years ago, I rashly predicted in this column that, despite all the bad publicity surrounding with-profits, the average policy was likely to continue to deliver better real returns than the average actively managed unit trust. Sadly, that assertion was made before the regulators decided to crack down on life company solvency after the Equitable Life debacle.

My original view was based on the actual, as opposed to perceived, performance of long term with-profits policies that were held to maturity (much better than most people realise), and the fact that the minority of investors who stay the course with their policies for a full term are in effect cross-subsidised by the many who fail to keep policies going. In a historical context, there was nothing much wrong in principle with owning a well-diversified long-term fund that "smoothes" investment returns over the years and provides life cover on top.

The problem with with-profits has turned out to be the high underlying charges in initial years, the false promises on which the idea was often sold and the inflexibility of the product as originally designed. Lack of full disclosure meant that few investors had a real idea of how effectively the policies were managed. Now you can re-create a with-profits policy of your own at relatively little cost, but that's a recent development.

Unfortunately, the regulators' action, which was made with exquisitely bad timing (just as the stock market was recovering from its worst bear market in a generation) has undermined my argument somewhat. One consequence has been a much wider dispersion in with-profits funds results, with well-funded life funds having much greater freedom to enjoy the strong performance of the stock market since 2003 than those less well-endowed.

Those funds with the freedom to invest a higher proportion in equities have naturally benefited from the high equity returns of the past four years, while others (including the rump of the Equitable Life fund) have been limited to investing in bonds and so condemned to low, single-digit returns that make meeting terminal values next to impossible in many cases.

The latest Money Management figures show how severely a combination of the 2000-03 bear market and the enforced move away from equities has hurt with-profits funds. The average pay-out on a typical ?50 per month 10-year policy has fallen from about ?10,000 in 1999 to ?7,000 now. On a typical 25-year policy, it has fallen from ?97,000 in 2001 to ?54,000. In both cases, however, the figure increased last year, for the first time since the bull market peak, perhaps marking a turning point in the level of terminal bonuses.

The funds that have provided the best pay-outs are mainly those of smaller insurance companies, often mutually owned, that were able to retain high equity weightings. The two that make the most appearances in the top 10 over the past decade, Liverpool Victoria and Wesleyan, both have about two-thirds of their fund invested in equities. The average is 38 per cent.

Over the long haul, the average surviving with-profits fund has produced a compound annual rate of return of 8.4 per cent over 25 years, 7.0 per cent over 20 years and 5.2 per cent over 15 years. This is slightly less than the 9.9 per cent, 8.3 per cent and 6.1 per cent a year achieved by the average all-equity unit trust/OEIC, but higher than the 6.9 per cent, 5.9 per cent and 5.2 per cent produced by the average balanced managed unit trust/OEIC, a more directly comparable animal.

Given that equities may well be close to the peak of the current cycle, this is not a bad turnout. The average real return on 25-year with-profits funds held to maturity has been 4 to 5 per cent, again a fair outcome - and the best funds have produced more; 7 per cent compound in real terms.

As always, the key is to pick your fund well: something I failed to do all those years ago, which is why surrendering remains a distinct possibility this year.

jd@independent-investor.com